Danon Disease Presenting with Slowly Progressive Cardiomyopathy and Harboring a Novel Missense Variant in the Lysosome-associated Membrane Protein Type 2 (LAMP-2) Gene

Abstract

Danon disease (DD) is a rare lysosomal storage disorder resulting from pathogenic variants of the lysosome-associated membrane protein type 2 (LAMP-2) gene. The disease is characterized by severe cardiomyopathy, which rapidly progresses to end-stage heart failure. This case, with DD caused by a missense variant, exhibited slow progressive cardiomyopathy and survived for an extended period despite being a male. A pathological analysis revealed that only a minority of the samples exhibited autophagic vacuoles with unique sarcolemmal features, which are typical of DD. Importantly, LAMP-2 expression was absent and the myocardial tissue contained a substantial amount of p62-positive aggregates.

Introduction

Danon disease (DD) is an extremely rare lysosomal storage disease that was first reported by Danon et al. in 1981 (1). In 2000, Nishino et al. discovered that the lysosome-associated membrane protein type 2 (LAMP-2) gene located at Xq24 is a causative gene for DD (2). A previous study suggested that LAMP-2 defects inhibit autophagosome-lysosome fusion, leading to inefficient lysosomal biogenesis and maturation (3). It is widely accepted that the mechanism underlying DD involves a functional defect in autophagy. DD shows an X-linked dominant inheritance pattern, and the typical clinical triad includes hypertrophic cardiomyopathy, skeletal myopathy, and mental retardation (4). In most cases, especially in male patients, cardiomyopathy progresses relatively rapidly at a young age, often requiring consideration for heart transplantation (5,6). Most LAMP-2 pathogenic variants identified to date are nonsense or frameshift variants predicted to truncate the protein (7). There are limited reports of DD cases with missense variants in LAMP-2 and there have been no reports that have studied skeletal and cardiac muscle tissues in detail. Although advanced research has been conducted, such as the establishment of induced pluripotent stem cells specific to DD (8), the pathology of this disease has not yet been completely elucidated and no treatment methods have been established.

Case Report

Clinical course

In 2008, a 33-year-old male patient was diagnosed with atrial flutter (AFL) with complete right bundle branch block. He was in good health and had no relevant medical history. A physical examination revealed no signs of heart failure or muscle weakness. Echocardiography revealed mild left ventricular hypertrophy (LVH) [interventricular septum thickness (IVST), 11 mm; posterior wall thickness (PWT), 11 mm; left ventricular end-diastolic diameter (LVEDD), 52 mm; and left ventricular end-systolic diameter (LVESD), 42 mm], partial aneurysmal architecture at the base of the posterior wall (Fig. 1), and a low left ventricular ejection fraction (LVEF) of 38%. The patient underwent catheter ablation for AFL, but no accessory pathways were found. Since LVEF recovered to 52% after 2 years, the differential diagnosis was considered to be arrhythmia-induced cardiomyopathy rather than myocardial disorders, such as sarcoidosis. However, in 2013 (age, 39 years), he developed a complete atrioventricular block (CAVB) and underwent pacemaker implantation. Subsequent follow-up was performed without the use of cardioprotective drugs, and his brain natriuretic peptide levels remained elevated at 300-500 pg/mL. Plain chest X-ray images showed that the cardiothoracic ratio increased over time. In 2015 (age 41 years), he suddenly experienced sensory aphasia and higher brain dysfunction (attention disorder, executive dysfunction, and memory disorder). A computed tomography scan revealed low-density areas in the left thalamus and left cerebellum, indicating a high possibility of cardiogenic cerebral infarction. As a result of rehabilitation, the patient did not have any serious neurological sequelae. Because the pacemaker leads were not compatible, we were unable to perform magnetic resonance imaging, which is necessary for the detailed investigation of cardiomyopathy and an evaluation of cerebral infarction.

Figure 1.

Clinical course, plain chest X-ray images, and echocardiographic findings. It was initially characterized by mild left ventricular hypertrophy and partial ventricular aneurysms; however, his slowly declined over a decade after catheter ablation and pacemaker implantation. The cardiothoracic ratio also increased over time on plain chest radiographs. CRTD: cardiac resynchronization therapy with defibrillator, LVEF: left ventricular ejection fraction, LVEDD: left ventricular end-diastolic diameter, LVESD: left ventricular end-systolic diameter

Thereafter, by 2020 (age 47 years), echocardiography showed a gradual decline in LVEF to 20%, enlargement of left ventricular diameter (LVEDD, 60 mm; LVESD, 55 mm), and a decrease in wall thickness (IVST, 9 mm; PWT, 7 mm). The patient developed congestive heart failure (CHF) for the first time and was hospitalized. He was started on treatment for CHF (including beta-blockers, angiotensin receptor neprilysin inhibitors, anti-aldosterone drugs, and sodium glucose cotransporter 2 inhibitors), and his pacemaker was upgraded to cardiac resynchronization therapy with a defibrillator. During the clinical course, his creatine kinase levels did not increase, and no skeletal muscle abnormalities, such as muscle atrophy or muscle weakness, were observed. The third edition of the Wechsler Adult Intelligence Scale showed no decline in intelligence quotient IQ (FIQ 93, VIQ 94, PIQ 92). The patient did not complain of either amblyopia or visual field defects, and no tests for retinopathy were performed. The patient's mother died of CHF in her mid-40s. His older brother had a pacemaker implanted in his 30s for CAVB and died suddenly at 42 years of age (Fig. 2A). His mother and brother died before DD was suspected and no detailed medical history was recorded. However, their activities of daily living were considered appropriate for his age, and there were no episodes of significant muscle weakness. Due to his family history, early onset cardiac conduction disturbance, and cardiac dysfunction, we suspected hereditary heart disease and performed detailed genetic testing and pathological evaluation.

Figure 2.

A pedigree and molecular genetic analysis. A: Pedigree of the family with Danon disease. Circle indicates female, and squares indicate male. Black symbols indicate affected individuals, and white symbols indicate unaffected individuals. The proband is indicated by an arrow. Slashed symbols indicate the deceased members. AFL: atrial flutter, CAVB: complete atrioventricular block, CHF: congestive heart failure, CRTD: cardiac resynchronization therapy with defibrillator, PMI: pacemaker implantation, SCD: sudden cardiac death. B: Deoxyribonucleic acid (DNA) sequence analysis of lysosome-associated membrane protein type 2 (LAMP-2) in the proband. A hemizygous nucleotide transition from T to A at nucleotide position 62 in LAMP-2 occurred in the proband. C: Electrophoresis of the reverse transcription polymerase chain reaction (RT-PCR) products from the proband and control using 1F and 2R primers. D: Direct sequencing of the RT-PCR fragment from the proband demonstrates c.62T>A substitution and no exon-skipping messenger ribonucleic acid (mRNA).

Molecular genetic analysis

A genetic analysis conformed to the principles outlined in the Declaration of Helsinki and was approved by the Ethics Committee for Medical Research of Kanazawa University. The patient provided written informed consent before participating in the study. Genomic deoxyribonucleic acid (DNA) was extracted from the peripheral blood leukocytes using standard methods. Targeted next-generation sequencing (NGS) analysis of the exonic regions of 180 arrhythmia- and cardiomyopathy-related genes of interest and DNA direct sequencing revealed a novel missense variant (c.62T＞A, p.Leu21Gln) in LAMP-2 gene was identified (Fig. 2B).

Because the nucleotide c.62T is located near the exon-intron boundary and the substitution of T to A may be predicted to affect messenger ribonucleic acid (mRNA) splicing, we analyzed peripheral blood mRNA. Total blood RNA was obtained using a standard protocol and reverse transcription polymerase chain reaction (RT-PCR) was conducted. Electrophoresis of the RT-PCR products from the proband showed a band of the same size as that of the wild type (Fig. 2C). Direct sequencing of the proband's band revealed a c.62T＞A missense variant and did not show any exon-skipping mRNAs (Fig. 2D).

Pathological examination

The pathological diagnosis of skeletal muscle specimens biopsied from the biceps brachii was performed at the National Center of Neurology and Psychiatry (Fig. 3). Pathological abnormalities in the skeletal muscle were limited to a small number of cells. Hematoxylin and Eosin (H&E) staining demonstrated sarcolemmal indentation and small basophilic granular structures in a few cytoplasms. The enzymatic activity of acetylcholinesterase has been found in the sarcolemma of meristematic fibers. One of these showed dot-like enzymatic activity, raising the possibility of an autophagic vacuoles with unique sarcolemmal features (AVSFs). No ragged red fibers or rimmed vacuoles were found on modified Gomori trichrome. Immunohistochemical staining revealed positive findings for sarcolemma proteins (including dystrophin, β-dystroglycan, merosin, and caveolin-3) within limited cytoplasm. Moreover, immunostaining revealed a complete absence of LAMP-2 expression.

Figure 3.

Pathological findings of a skeletal muscle biopsy. A: Hematoxylin and Eosin (H&E) staining showed a few fibers demonstrated sarcolemmal indentation and had small basophilic granular structures. B: Enzymatic activity of acetylcholinesterase (AChE) was found. C: On modified Gomori trichrome (mGT), no ragged red fibers or rimmed vacuoles were observed. Immunohistochemical staining revealed positive findings of β-dystroglycan (D) and dystrophin (E). F: Lysosome-associated membrane protein type 2 (LAMP-2) expression was completely absent.

Endomyocardial biopsy specimens were assessed at the National Cerebral and Cardiovascular Center (Fig. 4). H&E staining showed moderate-to-severe thickening of the endomyocardium, and Masson's trichrome staining showed interstitial fibrosis. Diastase-digested material positive for periodic acid-Schiff staining was locally identified and presumed to be glycogen. Immunostaining revealed complete lack of LAMP-2 expression. Many p62- and ubiquitin-positive aggregates were observed in the cytoplasm, thus suggesting the presence of chaperone-assisted selective autophagy defects. Electron microscopy revealed an increase in collagen fibers in the interstitium and proliferation of mitochondria between the myofibrils. Although not numerous, autophagic vacuoles with autophagic materials and glycogen granules were observed (Fig. 5).

Figure 4.

Pathological findings of an endomyocardial biopsy. A: Hematoxylin and Eosin (H&E) staining showed thickening of the endomyocardium. B: Masson’s trichrome staining showed interstitial fibrosis. C: Diastase-digested material positive for periodic acid-Schiff (PAS) staining was locally identified. D: Lysosome-associated membrane protein type 2 (LAMP-2) expression was completely absent. Many p62- (E) and ubiquitin-positive aggregates (F) were observed.

Figure 5.

Electron microscopic findings of endomyocardial biopsy. Electron microscopy revealed an increase in the number of collagen fibers and mitochondrial proliferation. Although not numerous, some autophagic vacuoles with autophagic materials and glycogen granules were observed. AVSF: autophagic vacuoles with unique sarcolemmal feature, MF: myofibrils, MT: mitochondria

Diagnosis of clinical DD

The guidelines of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) were followed (9), LAMP-2:p. Leu21Gln could be defined as “Likely pathogenic” based on one strong (PS3), one moderate (PM2), and one supporting (PP4). Hence, he can be diagnosed with clinical DD because he fulfills the following diagnostic criteria: 1) harboring a LAMP-2 mutation and 2) the presence of objective cardiovascular abnormalities such as mild LVH, reduced LVEF, AFL, and CAVB (10).

Discussion

The clinical phenotype of the present case was atypical for a male patient with DD. No mental retardation or skeletal muscle abnormalities were observed, and the clinical phenotype was limited to slowly progressing cardiomyopathy. Therefore, DD was not assumed to be a differential diagnosis based on the patient's initial clinical findings, and a definitive diagnosis was therefore delayed. The typical symptom onset is noted at an earlier age (13.3±8.0 years) in male patients (11).

Previous reports have shown various and a wide distribution of mutations in nine exons of LAMP-2 (12). No less than 110 different LAMP-2 mutations have been reported, most of which are nonsense or frameshift variants (7). In contrast, only nine missense variants in LAMP-2 were reported in the first male case in 2005 (7,13), and no missense variant with DD has been reported in Japan (12). We evaluated the missense variant, p.Leu21Gln identified in the present case using the ACMG/AMP guideline (9), and were able to classify it as “Likely pathogenic.” This variant was absent from controls in the Genome Aggregation Database and the comprehensive Japanese genetic variation database (TOGOVAR) (PM2). The proband's clinical phenotype and pathological findings are highly specific for DD, a monogenic lysosomal storage disorder (PP4). Immunohistochemistry for LAMP-2 and acetylcholine esterase staining directly from biopsied tissue from the proband provided strong evidence indicating that the function of the LAMP-2 protein is impaired (PS3). Recently, Abe et al. identified a pathogenic mutation in the noncanonical splice site of LAMP-2. This mutation induces exon skipping and structural changes in LAMP-2 (14). The missense variant identified in the present case was located near the exon-intron boundary. Although we considered the possibility that this might affect mRNA splicing, no exon skipping was observed. Most patients with missense variants are reported to present with mild cardiomyopathy, skeletal myopathy, and intellectual disability (15). Charron et al. performed genetic screening in patients with undiagnosed hypertrophic cardiomyopathy and reported that 4% had LAMP-2 mutations (16). DD due to LAMP-2 missense variants may go unnoticed as cardiomyopathy of unknown etiology, and NGS may provide a diagnostic stepping stone.

We herein report the detailed pathological findings of a patient with LAMP-2 missense variant for the first time. In the present case, the absolute number of AVSFs was limited, particularly in skeletal muscle. Nevertheless, LAMP-2 expression was completely absent in both the skeletal muscle and endomyocardium. A previously reported case with a missense mutation also lacked LAMP-2 expression in the muscle tissue (12), similar to the present case. It is difficult to explain the mechanism by which this missense mutation in the LAMP-2 gene induces a complete loss of protein expression. One possible reason may be that the missense variant causes conformational changes in the LAMP-2 protein, significantly reducing its affinity for the antibody. Further studies are needed to understand LAMP-2 expression in the skeletal and cardiac muscle tissues of patients with missense variant. Curiously, few p62- and ubiquitin-positive aggregates were discernible in the skeletal muscle; however, abundant accumulation was observed in the endomyocardium. This pathological finding is consistent with the clinical course of almost no skeletal myopathy. The number of AVSFs was not significantly increased in this missense variant compared to that in the nonsense or frameshift variants. This suggests that the LAMP-2 protein may not be completely absent and that the autophagy process might be partially compensated for by the function of defective LAMP-2 protein and other lysosomal membrane proteins. AVSFs, which are considered autolysosomes, are highly specific in the pathological diagnosis of DD (17). The number of AVSFs has been reported to increase with age (17). A previous report demonstrated that granules positive for lysosomal integral membrane protein-1, a lysosomal membrane protein, were observed in the muscle tissue of DD patients, consistent with the distribution of AVSF (17). This suggests that AVSF has lysosomal properties and may act as an autolysosome. However, autolysosomes with structural abnormalities due to LAMP-2 defects produce excessive reactive oxygen species and oxidative stress due to damaged organelles, mainly mitochondria, and toxic substances from cell debris. This is considered to be a potential molecular mechanism of DD (18). However, it remains unclear why AVSFs with such unique properties are formed, and how their accumulation is associated with muscle tissue degeneration.

Gene therapy for lysosomal diseases has progressed significantly in recent years. In 2020, the effectiveness of human LAMP-2B gene transfer in LAMP-2 knockout mice, a DD model, was demonstrated using recombinant adeno-associated virus 9 (AAV9.LAMP2B) harboring human LAMP-2B (19). LAMP-2 knockout mice treated with AAV9.LAMP2B showed a dose-dependent recovery of human LAMP-2B protein in the heart, liver, and skeletal muscle tissue, and no anti-LAMP-2 antibodies were detected. A phase 2 clinical trial for human applications is currently underway. As DD is a rare disease, the effectiveness and safety of gene therapy requires careful consideration. As gene therapy becomes more widespread in the future, early detection and intervention will become important regardless of the type of pathogenic variants.

Conclusion

DD due to missense variants is extremely rare and presents with a mild pathological phenotype. This should be considered when differentiating patients with cardiomyopathy of unknown etiology, and NGS is considered to be useful for early detection.

The authors confirm that written consent for submission and publication of this case report, including images and associated text, has been obtained from the patient in line with the COPE guidance.

The authors state that they have no Conflict of Interest (COI).

Financial Support

This work was supported by grants from a Grant-in-Aid for Scientific Research (C) from the Japan Society for the Promotion of Science (26460670 to K.H.), and the NOVARTIS Foundation (Japan) for the Promotion of Science to K.H.. The pathological diagnosis of skeletal muscle was supported by an Intramural Research Grant (2-5, 5-6) (PI: Ichizo Nishino) for Neurological and Psychiatric Disorders of the NCNP.